Network scheduler for real time applications

Abstract

A network scheduler for real time applications is described. The network scheduler supports scheduling of network packets in a multi-threaded operating system to insure correct scheduling order and to preclude kernel preemption of active threads. The network scheduler resides between a kernel (and device drivers) and multiple sessions at the user-level. The network scheduler uses input/output control primitives such as HOLD and RELEASE to control servicing of the queues of packets at the kernel or network device driver. The control primitives serve as a middleware interface to the underlying kernel or network device driver. The network scheduler functions as a proxy scheduler to manage the queues of packets at the kernel or network device driver.

What is aimed is:  
     
       1. A method of scheduling the use of a resource of a computer system, the method comprising: 
       receiving a scheduling constraint for a user-level thread; and  
       managing the scheduling of a corresponding kernel-level thread based on the scheduling constraint of the user-level thread  
       wherein the act of managing the scheduling of the kernel-level thread further comprises issuing one or more primitives to control scheduling of the kernel-level thread, the one or more primitives selected from the group consisting of: hold, release, check population of queue, check deadline of packet that is at the head of the queue, and check priority of packet that is at the head of the queue.  
     
     
       2. A computer-readable medium having computer-executable instructions stored thereon for executing a method of scheduling network packets, the method comprising: 
       holding a plurality network connection queues;  
       checking a deadline of a first one of the queues;  
       comparing the deadline of the first one of the queues to a deadline of a new network packet;  
       releasing the first one of the queues if the deadline of the first one of the queues is earlier than the deadline of the new network packet; and  
       assigning the new packet to a second one of the queues if the deadline of the first one of the queues is later than the deadline of the new network packet.  
     
     
       3. The computer-readable medium of  claim 2  further comprising checking the first one of the queues to determine if the first one of the queues is populated. 
     
     
       4. A computer-readable medium having computer-executable instructions stored thereon for executing a method of scheduling network packets, the method comprising: 
       holding a plurality network connection queues;  
       checking a priority of a first one of the queues;  
       comparing the priority of the first one of the queues to a priority of a new network packet;  
       releasing the first one of the queues if the priority of the first one of the queues is greater than the priority of the new network packet; and  
       assigning the new packet to a second one of the queues if the priority of the first one of the queues is less than the deadline of the new network packet.  
     
     
       5. The computer-readable medium of  claim 4  further comprising checking the first one of the queues to determine if the first one of the queues is populated.